Evaluation of early diagnostic approaches for malaria and pneumonia in children under-five presenting at the primary healthcare level in Benin City, Nigeria: a mixed methods study

Background Malaria and pneumonia are the leading causes of under-five mortality in sub-Saharan Africa especially in Nigeria. The Integrated Management of Childhood Illness (IMCI) guidelines were developed by the World Health Organisation (WHO)/United Nations Children’s Fund (UNICEF) as a strategy to reduce the burden of these and other preventable childhood diseases. However, there appears to be a paucity of evidence on the diagnostic performance of the revised IMCI guidelines and whether they offer an advantage over lay diagnosis (caregiver) for malaria and pneumonia management in Nigeria. Aim and specific objectives This study evaluates early diagnostic approaches (IMCI guidelines and lay diagnosis) for malaria and pneumonia in children under-five at the primary healthcare level. To address the overarching aim of the study, the following four specific objectives were studied: I. To assess the diagnostic accuracy of the IMCI guidelines and lay diagnosis (caregiver) for malaria and pneumonia in comparison to reference diagnostic approaches (microscopy and chest X-ray for malaria and pneumonia respectively). The extent of agreement between caregivers’ and health workers’ diagnosis of malaria and pneumonia is also assessed. II. To estimate the burden of malaria and pneumonia among children under-five presenting to study primary healthcare centres (PHCs) according to various diagnostic approaches. III. To determine the clinical outcomes in children diagnosed with malaria and pneumonia according to the IMCI guidelines and risk factors for severe outcomes. IV. To qualitatively explore caregivers’ and health professionals’ perspectives on lay diagnosis and IMCI guidelines as diagnostic approaches for childhood malaria and pneumonia. Methods A mixed methods approach was used for this study. The quantitative component used a consecutive sampling approach to recruit 903 children aged 2–59 months who met study eligibility criteria for malaria and pneumonia assessment according to the IMCI guidelines at presentation to five study PHCs in Benin City, Nigeria. Caregivers of these children were also asked what they thought the diagnosis was (lay diagnosis). Diagnostic accuracy was assessed in terms of sensitivity, specificity, positive and negative predictive values, Area under the Receiver Operating Characteristic Curves (AUROC) values and 95% Confidence Intervals (C.I). The extent of agreement was assessed in terms of Cohen’s kappa statistic (k) and 95% CI. The estimated burden of malaria and pneumonia during the study period was assessed using proportions and 95% C.I. Clinical outcomes in children diagnosed with malaria and pneumonia by the IMCI guidelines were described in terms of frequency and percentages, while the potential risk factors associated with clinical outcomes were assessed using odds ratios (ORs) and 95% C.I. For the qualitative component, health stakeholders (17 health professionals and 13 caregivers) who met the study eligibility criteria were purposively recruited and interviewed using semi-structured interviews. An inductive approach to thematic analysis was used for data analysis. Results Compared to microscopy, the diagnosis of malaria by health workers using the IMCI guidelines was poorly accurate with an AUROC value of 0.57 (with sensitivity and specificity of 51.8% and 61.3% respectively). Similarly, caregivers’ diagnosis of malaria was poor with an AUROC value of 0.55 (with sensitivity and specificity of 31.1% and 79.5% respectively) as compared to microscopy. Using the IMCI guidelines as the reference diagnostic test, caregivers’ diagnosis of malaria was more accurate (AUROC 0.60) in comparison to that of pneumonia (AUROC 0.54). There was a slight or minimal level of agreement (k=0.14; 95% CI: 0.09-0.19) between caregivers and health workers in the diagnosis of malaria and pneumonia. The estimated burden of malaria and pneumonia was relatively low, varying by the study local government areas, PHCs and seasonality, irrespective of the diagnostic approach. Where follow-up data were available, approximately 57% (172/304) and 78% (81/104) of the children diagnosed with malaria and pneumonia, respectively, recovered without complications within 30 days. Self-medication prior to presenting to study PHCs and use of preventive measures against malaria were independently and significantly associated with improved clinical outcomes. In contrast, exposure to solid fuels increased the odds of severe illness following malaria or pneumonia diagnosis. The qualitative component of the study found that caregivers rely on lay diagnosis despite the awareness of its limitations. The perceptions of malaria and pneumonia appeared to influence caregivers’ home management practices and health seeking behaviours. Caregivers showed willingness to be trained in the IMCI guidelines for improved home-based management of malaria and pneumonia. Health professionals believed that the IMCI guidelines were useful for managing both malaria and pneumonia. However, there are some recurring challenges to the wide-scale and sustainable implementation of the IMCI strategy in Nigeria. These include inaccurate diagnosis of malaria and inadequate funding. Conclusion The IMCI guidelines are crucial in the effective management (diagnosis and treatment) of malaria and pneumonia at the primary healthcare level in Nigeria. Although not perfect, lay diagnosis has an important contribution in the early detection and management of malaria and pneumonia at the community level in Nigeria. However, there is need for further investment in the training of both health professionals and caregivers in the IMCI guidelines for better health outcomes in under-five population. The training of caregivers in the IMCI guidelines and potential for a scale-up will benefit from careful design, piloting, implementation, and monitoring